Joint Spectrum Sensing And Access Algorithm In Cognitive Radio Networks

The subject of this thesis is supported by the National Natural Science Foundations of China,which are named Research on Key Technologies of Mobile Cognitive Radio Network(No.61271207)and Research on Key Technologies of Cognitive Wireless Sensor Network(No.61372104).By adopting relevant algorithms in machine learning and game theory,we solve the joint spectrum sensing and resource allocation problem in the presence of spectrum sensing data falsification(SSDF)attack.The main content is divided into two parts: malicious user detection based on support vector machine and game theory based joint spectrum sensing and resource allocation scheme.These two parts improve the robustness of the system and achieve a significant gain in resource allocation.In malicious detection algorithm,the paper adopts support vector machine(SVM)to analyse secondary user's behaviors from multi-round records of energy values and obtain the classification accuracy.In particular,the concepts of recognition probability and misclassification probability are introduced,and the tradeoff between misclassification probability and threshold of classification accuracy is theoretically obtained.Finally,we derive the asymptotic optimal property,which ensures the excellent adaptability under different SSDF attack strategies.In game theory based joint spectrum sensing and resource allocation scheme,the paper considers the sensing fairness of secondary users(SUs)and the interference constraint of primary user(PU),and we derive the decision threshold of SU.Besides,in resource allocation(RA)part,we model the RA as a mixed integer non-linear programming.This programming takes fully consideration of multi-user diversity,which improves the total access utilities of SUs.We decompose this programming into two parts: power allocation problem and timeslot-subcarrier assignment problem.The simulation results show that our proposed algorithm possesses the excellent performance in resource allocation.The thesis is divided into five chapters as follows.Chapter 1 is the introduction in which the background of the research work is described.Then the research content of the thesis isstated.In Chapter 2,the fundamental theory of support vector machine and game theory is introduced.We detail the optimal hyperplane and inner product kernel in support vector machine.Besides,we briefly describe the coalition game in characteristic form and coalition game in partition form.Chapter 3 proposes the malicious secondary user(MSU)detection algorithm based on SVM and we mainly focus on the SVM classification,the calculation of threshold in recognizing MSU and the derivation of asymptotic optimal property.Extensive simulation results demonstrate that our proposed scheme outperforms other existing approaches.In Chapter 4,we put forward a game theory based joint spectrum sensing and resource allocation scheme.This chapter mainly focuses on the spectrum sensing in a partition,joint timeslot-subcarrier and power optimization and partition iteration scheme.Finally,simulations are performed in the resource allocation result and the verification of SU's fairness.Chapter 5 is the summary of the research work including future research directions.